Background: Firearms are responsible for approximately 35,000 deaths per year (50% suicide, 50% homicide) in the United States. Although risk factors for firearms-carrying have been described, this knowledge has not yet translated into clinical screening tools.
Study Purpose: To develop a screening test for adolescent firearms-carrying.
Study Population: The 1999 National Youth Risk Behavior Survey (NYRBS), a cross-sectional survey of 15,000 high school students assessing risk-taking behavior.
Methods: Known risk factors for firearms-carrying contained in the NYRBS were candidates for two models predicting recent firearms-carrying. The “brief FiGHTS score” model, intended as a screening tool, included only terms not related to sexual behavior, significant substance abuse, or criminal behavior (Fi=fighting, G=gender, H=hurt while fighting, T=threatened, S=smoker). An “extended FiGHTS score” model, which included items on topics of substance abuse, sexual and criminal behavior, was developed for a more comprehensive interview.
Results: The brief FiGHTS score had an optimal sensitivity of 85% and specifity of 78% and an area under the ROC curve of 0.84. The extended FiGHTS score had an area under the ROC curve of 0.88 and predicted firearms-carrying well across all relevant subgroups in the data set, including age, gender, and regions of the United States. Similar results were obtained when the instruments were applied to the 1997 and 1998 NYRBS.
Conclusion: Both the brief and extended FiGHTS scores have high sensitivity and specifity for predicting firearms-carrying and should be evaluated prospectively.
Learning Objectives: N/A
Keywords: Firearms, Adolescents
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.